We know that perceptions of facts can be filtered by a strong partisan lens. Differing narratives regarding the recent election of Joe Biden and Kamala Harris have brought that into sharp relief. While a majority of Americans consider the issue settled, a surprisingly large minority does not. Based on a POLITICO/Morning Consult survey, fully 70 percent of Republican voters say they don’t believe the election was free or fair. But here is the interesting question: Is that 70 percent a real difference in perceiving the facts, or is it simply an expression of tribal identity? In other words, does this group truly believe the claims of “massive fraud,” or are they simply rooting for their team? Research conducted several years ago for the National Bureau of Economic Research (Bullock, et al., 2013) suggests that it is likely the latter.

The phrase “expressive response” refers to the observation that differing survey answers may not necessarily reflect differing views as much as they reflect differing motivations to align with one party over the other. “Partisan differences in survey responses may not solely indicate differences in true beliefs,” the authors write. “Instead, they may also reflect the expressive value of offering survey responses that portray one’s party in a favorable light.” So Republicans may simply read a question about electoral fraud, not as a question about a fact, but as a question asking them what side they are on, or which outcome they prefer. Of course, electoral issues are far from the only subject matter that might be distorted by the partisan lens: immigration, health care, national security, taxation, the economy, racial equality, and many other subjects will see a similar partisan influence when it comes to public opinion measurement. When these and other issues intersect with the sorts of questions you need to ask in voir dire, it will make sense at times to try to separate the partisan filter from the real perceptions. In this post, I’ll unpack this notion of “expressive response” and what it means for measuring attitudes in jury selections.

The Research:

The research team began with the observation of partisan differences regarding facts. For example, Republicans are more likely to say that budget deficits rose under Clinton, while Democrats are more likely to say that inflation rose under Reagan. Both claims are false, but the tendency for party members to “cheerlead,” or to support their own team is strong, so there are significant response differences based on party affiliation.

To uncover the extent to which these factual perceptions are real, they adapted an odd research method. They gave a small monetary incentive for correct answers, or for simply responding that they did not know the answer. The theory is that if party members truly have a different perception of the facts, then we would expect that the financial incentives wouldn’t make a difference. But if the partisan differences are just cheerleading, then it might be reduced when there is a cost to it.

And that is what they found. Adding a financial incentive for a correct answer reduced the partisan gap by about 55 percent. And the gap ended up being a whopping 80 percent smaller when they added the incentive for “I don’t know.” They concluded, “Apparent differences in factual beliefs between members of different parties may be more illusory than real.”

The Implications:

In voir dire, you usually aren’t asking factual questions, and you definitely aren’t giving a financial incentive. But the researchers’ implication still matters: We should not assume that every attitude with a partisan disconnect represents a true distortion of reality. “Partisans may disagree in surveys, but we should not take these differences at face value.” Instead, it is important to draw a distinction between belief on the one hand and affiliation on the other. For the attorney conducting voir dire and interpreting the responses, I think there are two implications.

Sometimes the Tribal Response Is Just What You Want

At times, it is helpful to know what “tribe” the potential jurors see themselves as belonging to. That is because their partisan identification is going to serve as a useful proxy for other issues that, we know, differ strongly by party. If your case ties into issues of personal responsibility, for example, we know that the ends of the political spectrum see that very differently, with conservatives hewing to a high expectation for rugged individualism, and liberals embracing collective responsibility and protection. When you aren’t able to ask that directly, you can often infer from their partisan leaning, by either looking at what they advertise on their own social media, or by asking a question that yields that partisan response. These days, that can mean anything from the election, to trust in courts and in medical authorities.

When It Isn’t, Be Sure to ‘De-Tribalize’ the Question

When you need to assess a potential juror’s true attitudes, you should take care in designing the question. For one thing, you don’t want to signal what is the correct or expected answer. But in these polarized times, you also don’t want to make it too obvious which answer aligns with which tribal identity. If, for example, attitudes toward immigration might bear on your case, then you don’t want to frame it as a question about “the President’s policies.” Instead, ask about hypothetical actions. For instance, you could ask, “Do you think that non-citizens should have access to U.S. courts?” You are still likely to have a partisan split in that answer, but you don’t want to give people the shortcut of seeing it as a question of which party they belong to.

In order to get a more honest response, you might also signal in your wording that there are reasonable people on both sides of an issue. If you are interested in personal responsibility attitudes in a products liability context, for example, you might emphasize that “Some people feel the government can and should be doing more to make sure products are safe, while others think the government should do less and let consumers make their own choices,” before asking which option they are closer to.

For the potential juror, answering is an act of self-presentation. In some situations, people will answer in ways that highlight their affiliations more than their honest opinions. Knowing that serves as a reminder to be careful that you are not signaling too much in the ways you phrase the questions.